Emerging data reveal excellent rates of adherence to antiretroviral therapy in sub-Saharan Africa - rates comparable to or surpassing those typically seen in the U.S. However, early successes may reflect a "honeymoon period," a time shortly after initiation of ARV therapy when very ill patients experience dramatic health gains and as a result, are strongly motivated to adhere. When the honeymoon period ends, adherence rates are likely to drop. Interventions to support adherence are needed in resource-limited settings, as they are in the U.S. However, interventions developed in this country cannot simply be transplanted to different environments. The first step in developing feasible, effective and sustainable interventions for resource-limited settings is to understand how these settings differ - culturally as well as socioeconomically - from resource-rich contexts and what the implications are for adherence. These understandings may then be translated to produce culturally appropriate theoretical models and interventions for adherence support. This two-year, qualitative, exploratory study builds upon ongoing NIH-funded quantitative research. It will provide new understandings of facilitators and barriers to adherence at a time when subsidized ARV therapy is becoming increasingly available in resource-limited settings. The study site is the HIV clinic at Mbarara University Teaching Hospital, Mbarara, Uganda. Data collection consists of multiple, in-depth interviews conducted at four time points over one year for each of 60 patients receiving ARVs through the clinic (Total Ints: 240). Observations of clinic activities and interviews with clinic staff will also be carried out. Grounded theory methods of category construction will be used to: (1) identify and represent facilitators and barriers to adherence, (2) characterize successful clinic-based adherence support efforts already in place, and (3) assess "degree-of-fit" for Uganda of a three-step model of adherence currently being used to guide intervention development in a very different resource-limited setting. Practical intervention strategies will be derived as the basis for intervention design and testing, which follow as next steps from this research. [unreadable] [unreadable] [unreadable]